The Midlife Crisis - My 833C Amp Build - Page 12 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 17th April 2013, 05:50 PM   #111
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by JamesHill View Post
I tried to use both capacitors and resistors on my 1250v Choke input (30H) power supply for my Push Pull GM70. I had massive ringing that I couldnt solve.

I gave up and used 866 MV's in full bridge which gave no such issues.

James.

Interesting! What type of rectifiers, standard or schottky? Si or SiC?

I went with the SiC schottky diodes because of their lack of a reverse recovery spike, which can kick off ringing.
  Reply With Quote
Old 17th April 2013, 06:26 PM   #112
diyAudio Member
 
Join Date: Nov 2003
Location: Bristol
Hi,

They were STTA512D.

Hopefully you will not have any such issues, but it may be worth testing for!

James
  Reply With Quote
Old 17th April 2013, 06:51 PM   #113
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by JamesHill View Post
Hi,

They were STTA512D.

Hopefully you will not have any such issues, but it may be worth testing for!

James
Ultrafast, Si diodes with a definite reverse recovery (albeit "soft"). Hopefully the use of SiC schottkys will prevent that issue

Definitely a watch-out, though.


Edit: Here's the datasheet for that diode. If you look on p.7 they have a chart showing the turn-off behavior with a serial inductance - it shows ringing.

http://www.datasheetcatalog.org/data...ics/mXqqsu.pdf

Last edited by Magz; 17th April 2013 at 07:02 PM.
  Reply With Quote
Old 19th April 2013, 08:29 PM   #114
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by Magz View Post


My inclination is to leave it the way I have it since the diodes are rated at 4800V and the caps at 5000V in a 2300V application. The diodes are all very close in Vf (which is inversely related to leakage current) and I've sorted them into quartets with Vf with .002V apart at worst (most are identical to 3 digits).

.

I've decided to add one more diode and cap per leg of the bridge, to give me a total of (5) 1200V diodes and (5) 1250V caps per leg, just to give me a little more headroom. I actually ordered 8 extra diodes to make the bridges for the driver circuits, and they're from the same batch as the others, so they should match pretty well.

Turns out Digikey is backordered until June for these diodes so I'm glad I ordered extra when I did, no telling how the next batch will match up with what I have. I don't need to worry about matching the ones in the driver bridges since they aren't in series.
  Reply With Quote
Old 27th April 2013, 09:17 PM   #115
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
I ordered and received 8 more 10nF caps so I've been able to finish the two HT rectifier bridges (see picture).

I tested them using a variac and dialed it up to 140VAC input, getting 166VDC output in both cases. Everything worked fine, so I let them warm up a bit then I measured the voltage drops across each diode in each leg of both bridges. For the worst matched leg, the voltages varied by 4.5% from highest to lowest - translated to 600VDC across each diode, which they would see if 3000VDC were across the leg, that makes 27V difference worst case. It looks like they balance just fine using the 10nF caps across them.

Interestingly, I used 5% caps, and the worst case imbalance was around 5%...

I will probably pot each bridge in epoxy to be sure nothing can ever short out between the wiring or between the solder pads; I may remove some of the pads between the highest voltage points anyway just to be safer. It shouldn't evolve too much heat at 160mA draw so I think that potting would be the safest way to go.


.
Attached Images
File Type: jpg 833 Amp HT Diode Bridge.jpg (366.4 KB, 706 views)

Last edited by Magz; 27th April 2013 at 09:23 PM.
  Reply With Quote
Old 27th April 2013, 09:47 PM   #116
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Quote:
Originally Posted by Magz View Post
I have read up on this for a few weeks now, and what I've found is:
Rather than reading multiple internet sources and trying to take the average, I suggest thinking about what you are trying to accomplish instead. If one source says 2+2 = 4 and two sources says 2+2 = 5, what is the value of 2+2 then? 4.66?

The issue with multiple diodes in series is that you have to distribute the reverse voltage evenly across the diodes. Failure to do this will result in uneven sharing of the voltage. If one diode ends up with a lion's share of the voltage, it's more likely to blow up due to reverse breakdown. This is what you are trying to avoid. The same is the case for capacitors in series. The only device I am aware of that can facilitate this charge/voltage sharing is a resistor. Hence, each diode will need a resistor in parallel. Pick a resistor of a high resistance to avoid dissipating excessive power in the resistor. You just need a little trickle current (higher than the expected leakage current of the diodes) to facilitate the charge/voltage sharing.

The caps across the diodes will not help in distributing the reverse voltage across the diodes. In fact, they are likely to make the problem worse. Capacitors cannot conduct DC current. The capacitors are used to tame the ringing caused by the turn-off transient of the diode resonating with the parasitic LC of the secondary circuit. In my experience, these caps do more harm than good if not selected properly.

Now, you can choose to ignore physics, oversize everything, do some hand-waiving, and hope for the best. But in a $10k amplifier build operating at 2500 V with more than enough energy to kill you twice without breaking a sweat, I think you need to slow down and understand the problem rather than relying on a plug-n-play approach.

It's good to see that you are testing things at lower voltages when possible. However, note that leakage currents in the diodes is rather non-linear and very temperature dependent. In addition, you can't rely on the matching between diodes. Dude! You're playing with fire here...

I don't mean to be nasty, but this is serious stuff. 2+2 = 4 not 4.66!

~Tom
__________________
21st Century Maida Regulator, Universal Filament Regulator, Damn Good 300B, Novar Spud, and more...
Neurochrome : : Audio - http://www.neurochrome.com/audio - Engineering : : Done : : Right

Last edited by tomchr; 27th April 2013 at 09:52 PM.
  Reply With Quote
Old 27th April 2013, 10:11 PM   #117
rmb is offline rmb  United States
diyAudio Member
 
Join Date: Nov 2012
Location: western PA
Mags: Wire 1 megohm 2 watt wirewound resistors across each diode in each stack. That will distribute the voltage across each diode.
  Reply With Quote
Old 27th April 2013, 10:15 PM   #118
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
My experiment above seems to show they share voltage pretty well, at least at 166VDC. I didn't want to go too much higher than that and be probing around every diode with a multimeter - too much chance for a slip. Yes, I understand that leakage can be non-linear, but really how much more can that vary within the same diode batch? Probably an unanswerable question.

The problem I have with resistors is that from what I read they are usually the component that fails in these types of series bridges. I'll need to get high voltage ones for sure if I add them. Maybe I'll add some and see; I'll have to think about it.


.

Last edited by Magz; 27th April 2013 at 10:19 PM.
  Reply With Quote
Old 27th April 2013, 10:25 PM   #119
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by rmb View Post
Mags: Wire 1 megohm 2 watt wirewound resistors across each diode in each stack. That will distribute the voltage across each diode.
Yeah, that's what I hear. I still may just because some folks seem so adamant about it, but I was curious as to see how well it would work with just the caps. Works pretty well, at lower voltages at least.
  Reply With Quote
Old 27th April 2013, 10:43 PM   #120
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by tomchr View Post

It's good to see that you are testing things at lower voltages when possible. However, note that leakage currents in the diodes is rather non-linear and very temperature dependent.
~Tom
Interestingly, the datasheet from Cree shows that the Ir is quite steady with voltage and temperature up to at least 1000V per diode and 125 degC. Must be the SiC.
Attached Images
File Type: jpg Cree SiC Schottky 1200V Ir.jpg (75.0 KB, 663 views)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
mono block amps with 2 x 833C overkill ? even possible to DIY? Thesoundmen Tubes / Valves 25 9th June 2014 02:27 AM
Good Source for Nice Analog Panel Ammeters for 833C Build? Magz Tubes / Valves 12 7th March 2013 07:49 PM
Small and portable headphone amp Crisis firefart_1st Headphone Systems 15 26th May 2010 08:24 PM
Midrange + midlife crisis ronzeman Multi-Way 7 1st September 2008 02:39 PM
No More Sound Crisis john2890 Everything Else 1 23rd December 2005 06:50 AM


New To Site? Need Help?

All times are GMT. The time now is 11:10 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2